Chat Stateful Streaming

ARK Platform Example / Chat Stateful Streaming

• Copy

  
  import openai
  import uuid
  
  ark_api_key = "API_KEY"
  ark_base_url = "https://api.ark-labs.cloud/api/v1"
  model_name = "gpt-4o"
  
  client = openai.OpenAI(api_key=ark_api_key, base_url=ark_base_url)
  
  # generate a conversation/session UUID once and reuse it for all requests
  ark_session_id = str(uuid.uuid4())
  
  def stream_completion(messages):
      response = client.chat.completions.create(
          model=model_name,
          messages=messages,
          extra_body={"ark_stateful": ark_session_id},  # enable stateful mode, using the same UUID for all requests
          stream=True,
          max_tokens=8000,
      )
      for chunk in response:
          delta = chunk.choices[0].delta
          if delta and delta.content:
              print(delta.content, end="", flush=True)
      print()
  
  first_messages = [
      {"role": "system", "content": "You are a helpful assistant."},
      {
          "role": "user",
          "content": (
              "Tell me a story about a brave knight traversing space in a small rocket "
              "who's lost because GPS only works on Earth. 200 words."
          ),
      },
  ]
  print("\nFirst response:")
  stream_completion(first_messages)
  
  second_messages = [
      {"role": "system", "content": "You are a helpful assistant."},
      # no need to resend history, ark_stateful mode remembers it
      {"role": "user", "content": "Translate the story to German, please."},
  ]
  print("\nSecond response:")
  stream_completion(second_messages)